Definitive Demand D11 Speaker Review

[KaiserSoze]

A whirly bird thingymajiggy, doomaflooper scoobydoo.

That's exactly what I was going to say.

 

[KaiserSoze]

Like these:
View attachment 80029

View attachment 80030

The swirl panels reminded me of those little spin toys we used to play with under the fluorescent lights. The fluorescent lights have a strobe effect that caused the spinning top with swirl pattern to appear to be spinning slow in one direction then slow to a stop and reverse direction, speed up then slow down and reverse direction again. The aliasing effect, you know. Same as the wagon wheels on TV and in the movies.

 

[BostonJack]

Yah and also the one on the MM de Capo Reference 3a Be, is pretty unusual.
I think both are hoping to draw attention to what is otherwise not interesting. Sales and or maybe they actually have developed a useful design ala KEF tangerine tweeter guides.

View attachment 80016View attachment 80017

This whacky little thing looks like an acoustic Fresnel lense. Based on that scale, my guess is that it is not effective as such as the geometry is just way too small. Maybe its just a bit of technical jewelry?

 

[ROOSKIE]

This whacky little thing looks like an acoustic Fresnel lense. Based on that scale, my guess is that it is not effective as such as the geometry is just way too small. Maybe its just a bit of technical jewelry?

Yah, I suspect likely more to drive sales than sound. That said I think that that speaker (the MM de Capo) uses shallow slops on the X-over and as such the woofer may contribute to the sound well in the range that a phase plug could truly be useful. Not that it needs to look so crazy but really with all competition these days anything to stand out is a temptation for a company.
I actually like the look of the over-all design on both the Def Tech D11 and the Reference 3a. I like a lot of looks though, running the range from bare mdf and plywood to old school JBL retro and speakers with exotic and beautiful hardwoods to heavy industrial of the KEF reference and others. Basically I like speakers.

 

[watchnerd]

Yah and also the one on the MM de Capo Reference 3a Be, is pretty unusual.
I think both are hoping to draw attention to what is otherwise not interesting. Sales and or maybe they actually have developed a useful design ala KEF tangerine tweeter guides.

View attachment 80016View attachment 80017

Now I want to upgrade my speaker with gold screws.

 

[KaiserSoze]

Sorry if my post wasn't clear, as the concept is really quite simple. I will try to clarify.

When tweeters are offset, each half of the speaker will radiate sound differently, and one side of the speaker's radiation will almost inevitably be better than the other. This is certainly the case for D11, as shown in my earlier post, as well as the D9 I'd tested. You, therefore, want to be careful about how you position the speakers.

Problem is, the spinorama weighs both sides of the speaker equally, probably because most speakers are indeed symmetrical. But my anecdotal experience suggests that whichever side of the speaker is closest to your sidewall will be perceptually dominant -- which I believe is an intuitive result. If the good side is closer to your sidewall, it will have a louder and earlier reflection.

Deftech clearly designed the speakers with this in mind, and it's why the speakers are specifically labeled as 'Left' and 'Right.'

I made a drawing to better illustrate what I mean:

View attachment 80032

Because the offset is mirrored, you get the same perceptual 'weighting' to the good side with the left speaker as well.

Anecdotally, the advantage of this approach was very obvious to me during testing. When I tried switching the D9 to the opposite side of what Deftech recommended, it was like a different speaker; timbrally similar, but the soundstage had compressed majorly and become relatively 'fuzzy'. (You could of course argue confirmation bias, but that's a risk we take with any sighted listening.). I also tried positioning the speakers a little bit wider to compensate for the difference in the tweeter-to-tweeter width, to no avail.

Anyway, all this is to say that I expect D11 sounds (at least a little) better than the spinorama and preference score suggest, but the reasons why are still apparent from the detailed measurements. You can of course achieve good directivity without an offset tweeter, but I believe the approach here is effective too (at least for the D9). A waveguide will usually narrow directivity, and a slightly rounded edge will only do so much. The D9 illustrates how this can work more clearly than the D11. Its 'good side', for example, had some pretty excellent results for a simple cone and dome. The 'bad' side on the other hand, was awful, but it's not what my ears picked up on... except when I inverted the setup.

As for whether the DBR-62 would be better, I agree that it's probably the better speaker, but I'm not convinced it's the better speaker. Its on-axis is kind of janky, and though it doesn't have as sharp of resonance at 600Hz, it still has a sizeable one that is wider. On-axis and PIR of the D11 and the Elac:

View attachment 80040

As for the design, the Elac is definitely closer to my personal aesthetic actually. But the deftech is really built on the exterior like a fairly high end speaker -- the aluminum faceplate is rare and it just feels all fancy - higher-end than say, the Focal Chora or Buchardt S400. It feels just as well built as the KEF R3 imo. From what I can tell the elac looks pretty nice but the finish is fairly cheap. Of course, this is all very subjective.

There are pretty much always going to be better options for cheaper of course, but I thought I'd illustrate why the speakers might not be horrendous. My pick for the price would be the Focal Chora 806, but they're definitely a step down in build quality (though more to my aesthetic).

That helps to explain some of what you had previously written. But you had written "Not a big deal as the speaker is specifically designed to be listened off-axis". First off I couldn't figure out exactly what it was that you decided was "Not a big deal." And then I wondered why you decided that Definitive intended for the speaker to be listened to off-axis, which struck me as sort of odd. I may have deleted the post where I started to ask you about this specifically, I don't remember whether I did. But if you only mean that Definitive recommends one specific side for the side closest to the wall, that is not entirely the same thing as listening off-axis. And maybe I already mentioned this, but when two speakers are toed in, this is to allow them to be listened to on-axis. If the speakers are aimed straight to the front and parallel to each other, there is no location in the room where both speakers can be listened to on-axis. So if Definitive actually did say that the speakers should be listened to off-axis, maybe this only means that you shouldn't toe them in.

My curiosity having gotten the best of me, I went to Definitive's site and downloaded the manual. Here's what I found: "Toeing the speakers in slightly, if possible, toward your listening area will also provide a stronger center image." The usual reason for the recommendation for toeing speakers in slightly is so that a listener sitting in the sweet spot at the middle will be more nearly on-axis, compared to what the situation would be if the speakers were aimed straight to the front and parallel. As such, I interpret this to be just the usual recommendation that the speakers should ideally be listened to on-axis to the extent practical. I don't see any reason to infer or conclude that Definitive recommends listening to these speakers off axis, and certainly no reason to think that they were designed with the express intent that they be listened to off-axis.

As for the rationale for the funky waveguide, on Definitive's site I found this: "Offsetting the tweeter by 5° delivers better high-frequency dispersion for a more precise center stereo image by eliminating undesirable symmetric diffraction off the corners of the front baffle." Until I read this I did not realize that the tweeters were angled slightly off from directly ahead (by 5 degrees evidently) in addition to the off-center placement. They had to go and obfuscate things by writing "a more precise center stereo image", which is just plain old marketing spiel, but it is apparent from the rest of the sentence that the goal of this was the fairly common goal of mitigating baffle edge diffraction. They took it a step further by angling the tweeter. Maybe this yielded significant improvement, but it would be difficult to say. In any case I do not find any reason to infer that Definitive recommends listening to these speakers slightly off-axis, and certainly nothing that would suggest to me that these speakers were designed with the express intent that they should be listened to slightly off-axis.

 

[watchnerd]

They took it a step further by angling the tweeter. Maybe this yielded significant improvement, but it would be difficult to say.

In a crowded market, you need a unique selling point.

Angling the tweeter is "unique" and probably doesn't increase the BOM much to impact margins.

Fits with the gimmicky-looking phase plug.

 

[KaiserSoze]

In a crowded market, you need a unique selling point.

Angling the tweeter is "unique" and probably doesn't increase the BOM much to impact margins.

Fits with the gimmicky-looking phase plug.

It seems to me that this speaker has a lot that amounts to plain old marketing gimmickry. As I previously mentioned I have been curious about Definitive's dual-surround drivers for some time, and have wondered whether the rationale for this is that it eliminates the need for a spider and thereby improves the linearity of the suspension's "restoring force". Eliminating the spider and replacing it with a second surround would likely have this effect, because as compared to spiders, surrounds typically behave more like an ideal spring (i.e., a spring that perfectly obeys Hooke's law, where stiffness is a linear function of displacement). Somewhere in one of Klippel's papers there is a graph that illustrates this. So I thought that perhaps this was the explanation for the dual surround. But I haven't seen any images of the drivers that would permit me to tell whether there is or isn't a spider. And if this was the goal, to improve the linearity of Kms, it doesn't seem to have had the intended effect given the distortion levels seen in Amir's measurements, which frankly are some of the worst we've seen in the various speakers he has measured of late. If the dual surround isn't intended to reduce distortion, then I can't fathom what the purpose would be, because it doesn't seem likely that it would improve the frequency response. I had previously thought it likely that these drivers are low in distortion as a result of the dual surround, but now I'm more inclined to think that the dual surround is just another gimmick.

 

[amirm]

wondering if you ever listen first and then take measurements. any chance you could be biased by your measurements?

I did that exact thing in this instance:

Speaker Subjective Listening Test
I listened to the D11 without seeing the measurements.

I have done it a few times and it has not made a difference. In general when I do make (compute) the measurements (which I did NOT do in this instance), there is a long enough time gap that I don't remember exactly what they said before I listened. I only pull up the measurements after I have listened to see if it agrees with my impression. I then confirm the same with application of EQ. A lot of times I abandon EQ as I did here when the frequency response flaw did not represent what I was hearing.

 

[amirm]

Adding on to my last post, unlike people who make manual measurements of speakers where they clearly see the frequency response as they measure at different angles, with Klipple only raw data is generated. Computation is needed to generate the final response. Even if I look at this response, it does not get ingrained in my mind the same way as if I made the DIY type measurements where you see over and over again how the speaker measures.

 

[amirm]

Someone asked about near-field measurement distance. Klippel recommends setting the tweeter point at about 1 centimeter from the driver. I do that and make the tweeter near-field measurement. I then maintain the same distance and only change the Z axis to center of the woofer. I used to move it closer to woofer but I no longer do that. It was dangerous as if I forgot and moved the Z axis, it would hit the woofer surround!

 

[amirm]

Someone asked about rolling off very low bass. I did try that with EQ. In the few parameters I tested, it did some good on some tracks but on others, it lowed the bass subjective so I abandoned it.

In general if I don't achieve success in about 10 to 15 minutes with manual EQ (eyeballing), I give up.

One of you should build a web service to auto-generate the EQ settings given an on-axis graph. That way I can use it to generate a quick EQ to test.

 

[ROOSKIE]

My curiosity having gotten the best of me, I went to Definitive's site and downloaded the manual. Here's what I found: "Toeing the speakers in slightly, if possible, toward your listening area will also provide a stronger center image." The usual reason for the recommendation for toeing speakers in slightly is so that a listener sitting in the sweet spot at the middle will be more nearly on-axis, compared to what the situation would be if the speakers were aimed straight to the front and parallel. As such, I interpret this to be just the usual recommendation that the speakers should ideally be listened to on-axis to the extent practical. I don't see any reason to infer or conclude that Definitive recommends listening to these speakers off axis, and certainly no reason to think that they were designed with the express intent that they be listened to off-axis.

This comes up a lot here. It seems like folks really expect the manual do be more than it is - something that is never read. I have to agree with the assessment that Definitive is unlikely to designed to be a fully toed in speaker purely based on intended market and home environment norms.
Lots of speakers are not designed this way but the manual is not explicit (about anything)
Andrew Jones himself said he never listens for enjoyment on axis and that that is especially true due to symmetrical L-R driver placements and diffraction. Therefore I'd suggest he doesn't really design for it but this is not in the manual, only found in an interview.

I really don't think the typical Best Buy/Crutchfield shopper is going to use toe, (maybe a little - but not fully on axis). These speakers main market (in the USA) is the typical Best Buy and Crutchfield shopper who is splurging a little.
I especially consider these are likely to end up in a home theater set up with a center channel. Toe in there is even less likely and needed.

I really don't think we should be reading the product manuals in this site for information about set-up any more than we would use the specifications published in those manuals as a substitute for the Kipple.
That is not helpful for the less experienced who come here looking for that one day or two of insight while they set-up their gear. I hope we can help them set-up correctly for the best sound regardless of what is in literature or manufacturers manuals.

IMHO even as someone who's music only stereo is center piece in my living room (no TV in this house) adjusting toe to taste is a part of the set-up. I am fine with a speaker toed completely on axis and I am fine with it straight ahead. Whatever is best. I am simple hoping to capture the best synergy with the particular speaker and the room and my personal fine tuning to taste.

When I get these at home I will deff (pun intended), play with the toe and subjectively asses. I can also take some in-room measurements at some point.

*As a side note look at the Infinity R162 and the RC263 (as measured on this site). See how the treble on axis is elevated in the R162 compared with the RC263. The off axis treble in the R162 very closely matches the RC263. These match well when the R162, being used as a L+R is not toed in very much and of course the center is completely on axis.
I found this to me to be a very clear example of Infinity's understanding of the actual common household usage and it is not mentioned in manual.

 

[MZKM]

One of you should build a web service to auto-generate the EQ settings given an on-axis graph. That way I can use it to generate a quick EQ to test.

Listening Window may be more useful.

No way in my abilities, but it would be interesting if someone could program an auto-EQ (10-band limit to make it more universally implementable) that achieves the highest preference rating. This of course would be very challenging.

 

[spacevector]

One of you should build a web service to auto-generate the EQ settings given an on-axis graph. That way I can use it to generate a quick EQ to test.

Amir, would you be open to hosting this on ASR? I think @Maiky76 has a script that spits out EQ when fed the spinorama. I'm not sure how it will be adapted to a webpage but it should not be a huge amount of work. @Maiky76 - have you already published your program? I know you published the Spinorama score calculator, not sure about AutoEQ (TM) .

 

[KaiserSoze]

This comes up a lot here. It seems like folks really expect the manual do be more than it is - something that is never read. I have to agree with the assessment that Definitive is unlikely to designed to be a fully toed in speaker purely based on intended market and home environment norms.
Lots of speakers are not designed this way but the manual is not explicit (about anything)
Andrew Jones himself said he never listens for enjoyment on axis and that that is especially true due to symmetrical L-R driver placements and diffraction. Therefore I'd suggest he doesn't really design for it but this is not in the manual, only found in an interview.

I really don't think the typical Best Buy/Crutchfield shopper is going to use toe, (maybe a little - but not fully on axis). These speakers main market (in the USA) is the typical Best Buy and Crutchfield shopper who is splurging a little.
I especially consider these are likely to end up in a home theater set up with a center channel. Toe in there is even less likely and needed.

I really don't think we should be reading the product manuals in this site for information about set-up any more than we would use the specifications published in those manuals as a substitute for the Kipple.
That is not helpful for the less experienced who come here looking for that one day or two of insight while they set-up their gear. I hope we can help them set-up correctly for the best sound regardless of what is in literature or manufacturers manuals.

IMHO even as someone who's music only stereo is center piece in my living room (no TV in this house) adjusting toe to taste is a part of the set-up. I am fine with a speaker toed completely on axis and I am fine with it straight ahead. Whatever is best. I am simple hoping to capture the best synergy with the particular speaker and the room and my personal fine tuning to taste.

When I get these at home I will deff (pun intended), play with the toe and subjectively asses. I can also take some in-room measurements at some point.

*As a side note look at the Infinity R162 and the RC263 (as measured on this site). See how the treble on axis is elevated in the R162 compared with the RC263. The off axis treble in the R162 very closely matches the RC263. These match well when the R162, being used as a L+R is not toed in very much and of course the center is completely on axis.
I found this to me to be a very clear example of Infinity's understanding of the actual common household usage and it is not mentioned in manual.

Okay, but honestly I was not interested in the question of whether it is advisable to toe in the speakers, or for that matter the question of whether it is advisable to listen to these speakers off-axis. I was only interested in knowing whether certain claims that were previously made, to the effect that Definitive had designed these speakers expressly to be listened to off-axis and recommended listening to them off-axis, were based in reality.

My reason for doubting these claims was much simpler than the reason you gave for why you agreed that it was unlikely that Definitive had designed these speakers with the express intent that they be listened to off-axis. My reason was simply that it struck me as something that most manufacturers wouldn't be inclined to do, and more the kind of thing that people pull out of thin air and then post on a Web forum. Your reason for agreeing that is unlikely that Definitive had designed these speakers with this express intent was "purely based on intended market and home environment norms". To your way of thinking Definitive would not have designed these speakers this way because the class of consumer to which they were intended to be sold lacked the sophistication that would be needed before it would make sense for a manufacturer to design and sell speakers expressly intended to be listened to off-axis. When I first read that it struck me as sort of pretentious, but after thinking some more about this your reasoning seems perfectly reasonable to me. Which is partly why I wanted to edit this post (you are reading the edited version).

 

[amirm]

Amir, would you be open to hosting this on ASR?

Yes although someone needs to advise as to how secure the service is.

 

[ROOSKIE]

Since you wrote this as a reply to my post, it seems appropriate for me to point out that I was not asking whether it is or isn't a good idea to toe in the speakers (or to listen off-axis). I have no interest in this particular question, for reasons to include the fact that the only answers would be other people's opinions. But no worries on this though, because it was an easy misunderstanding for anyone to have made. If you have any sincere interest in correctly understanding the questions I was addressing, you may be able to satisfy that interest by reading my post again. But I expect not.

You wrote, among other things, that you agreed that it is unlikely that Definitive had designed these speakers to be "fully toed in". The reason you gave for why you decided that you were in agreement on this question is "based on intended market and home environment norms". The actual question was not per se with whether they had been designed to be "fully toed in", but rather with whether they had been designed to be listened to off-axis. The two things are very closely related of course, but I point out the distinction because I think it is an important distinction in spite of the obvious relatedness, and also because I want to share my reason for doubting that Definitive had designed the speakers with the intent that they be listened to off-axis. My reason was that this simply struck me as a very odd thing for any manufacturer to do, and instead struck me as very much the kind of thing that people pull out of thin air and then post on a Web forum where the notion is often accepted at face value by lots of sheep-people. If I ever get around to writing that book explaining how the Internet and the Web actually work, I think I'll include a chapter on this phenomenon.

Yah man, I get you.
I understand your doing your thing and I am doing mine. Investigating the speakers.

You have to understand that I'm not just replying to you - I am discussing this stuff in a more general sense for the public forum. I am not misunderstanding you here friend. One thing to note is that the measurements here are always on axis (which would equate with a full toes in) and so I do like to mention from time to time that may not represent the design goals. (even if unfortunately the manual does not clarify them)

You are kind of restating you general case and I will as well. I really don't agree that it is a weird thing for a manufacturer to design a speaker that is meant for off axis listening. I suspect it is actually quite common. Many people will never place the speakers anywhere but straight ahead for many reasons and there is not really a great reason to design a speaker for on axis listening over off axis or even way off axis (zero toe). In fact there are many benefits to designing for off axis including mitigating diffraction effects. (there are also many benefits to an on axis design of course)
Dali is one manufacturer of note who at least lets people know this. They design many of their models for no toe in.

Anyway
In any case I do think it good to be thorough and have these conversations.

As said in his orignal @napilopez said "Anyway, this speaker is designed to be used with the offset tweeters on the outside edges, and appears be designed for use with minimal toe in (manual says: "Toeing the speakers in slightly, if possible, toward your listening area will also provide a stronger center image.) That's how the D9 sounded best. "

I can see you value toughness and I mean no harm but you followed by replying to him with the same quote from the Deff Tech manual.

Anyway, I will be interested to play around with these somewhat odd speakers and I most deff think you nailed the passive radiator resonances due to box dimensions. I am going to take it apart and see what can be done there to confirm.

 

[Maiky76]

Hi,

I am glad that I got the ball rolling with regards to the issue 500 - 700Hz.
I have formulated one hypothesis, some people have questioned it, that's the point and the more the merrier!
It would be great to find an explanation that is consistent with the observation...

After thinking about it, indeed the Passive Radiator looks suspicious when compared with the S400.
Also as I mentioned internal resonance are not 100% ruled out yet.

@amirm let's exploit the NFS data.
Could you publish the balloon directivity diagram for the 536Hz and 615Hz (or in the vicinity)?
It might help us understanding the phenomenon.
It seems that the resonance at 536Hz out of phase @0deg becomes in phase at +70 down to -70deg

Can @ROOSKIE measure the surround of the PR if you can?
From the pictures you could also measure the thickness of the walls as well.

With regards to EQ, I don't plan to publish my GA optimizer.
I am sure the clever guy at Vituix or others can figure out what to do.

Attach files

 

[napilopez]

That helps to explain some of what you had previously written. But you had written "Not a big deal as the speaker is specifically designed to be listened off-axis". First off I couldn't figure out exactly what it was that you decided was "Not a big deal." And then I wondered why you decided that Definitive intended for the speaker to be listened to off-axis, which struck me as sort of odd. I may have deleted the post where I started to ask you about this specifically, I don't remember whether I did. But if you only mean that Definitive recommends one specific side for the side closest to the wall, that is not entirely the same thing as listening off-axis. And maybe I already mentioned this, but when two speakers are toed in, this is to allow them to be listened to on-axis. If the speakers are aimed straight to the front and parallel to each other, there is no location in the room where both speakers can be listened to on-axis. So if Definitive actually did say that the speakers should be listened to off-axis, maybe this only means that you shouldn't toe them in.

My curiosity having gotten the best of me, I went to Definitive's site and downloaded the manual. Here's what I found: "Toeing the speakers in slightly, if possible, toward your listening area will also provide a stronger center image." The usual reason for the recommendation for toeing speakers in slightly is so that a listener sitting in the sweet spot at the middle will be more nearly on-axis, compared to what the situation would be if the speakers were aimed straight to the front and parallel. As such, I interpret this to be just the usual recommendation that the speakers should ideally be listened to on-axis to the extent practical. I don't see any reason to infer or conclude that Definitive recommends listening to these speakers off axis, and certainly no reason to think that they were designed with the express intent that they be listened to off-axis.

As for the rationale for the funky waveguide, on Definitive's site I found this: "Offsetting the tweeter by 5° delivers better high-frequency dispersion for a more precise center stereo image by eliminating undesirable symmetric diffraction off the corners of the front baffle." Until I read this I did not realize that the tweeters were angled slightly off from directly ahead (by 5 degrees evidently) in addition to the off-center placement. They had to go and obfuscate things by writing "a more precise center stereo image", which is just plain old marketing spiel, but it is apparent from the rest of the sentence that the goal of this was the fairly common goal of mitigating baffle edge diffraction. They took it a step further by angling the tweeter. Maybe this yielded significant improvement, but it would be difficult to say. In any case I do not find any reason to infer that Definitive recommends listening to these speakers slightly off-axis, and certainly nothing that would suggest to me that these speakers were designed with the express intent that they should be listened to slightly off-axis.

You're definitely right that I asserted that with too much confidence that the speakers are meant to listen to off-axis, so I apologize about that. DefTech isn't explicit about it, it's just how I interpreted the text and illustration of the manual. To me "toe the speakers slightly" means "you don't need to toe them in all the way." Otherwise they could just say aim to aim them at the listening position as others do, but @ROOSKIE is right that we sholdn't look at manuals too closely for advice unless they are explicit about intentions (such as Neumann, which tells you very specifically where the reference axis is, down to the millimeter). I can see why you might've interpreted that differently.

However, please keep in mind my off-axis comment was a minor point in my original post. It was primarily meant to suggest the on-axis hot treble of the D9 (which is not present in Amir's measurements of the D11 anyway) probably isn't a huge issue in practice. The bit about listening off-axis had little to do with my main point about why the measurements of the D11 are better than they appear at first glance, and the measurements I posted are all referenced to the on-axis anyway. So my larger point about the use of offset tweeters and how the close-wall reflections are perceptually dominant remain.

But as we're on the topic... I do think it's totally normal for speakers to be designed with listening off-axis in mind. So I shouldn't have said "specifically designed to be listened off-axis ," but I think my later assertment that it "appears to be designed for use with minimal toe-in" was not unreasonable.

Dennis Murphy has mentioned doing so for the BMRs, Buchardt explicitly mentions the S400 and A500 are designed to listened with no toe-in (with the default tuning for the latter), Andrew Jones does it, Dali does it, and different folks at KEF have told me three times that it designs speakers to be listened to off-axis as well (but you have the choice).

It is also worth noting that in the only study that I know that analyzes the typical positioning of speakers in home listening environments(devantier, 2002), the majority of listeners had their speakers set up for off-axis listening:

So I don't think the idea that engineers making HiFi speakers are designing them to be listened to off-axis is unreasonable.